Inhibition of Cereal Rust Fungi by Both Class I and II Defensins Derived from the ﬂowers of Nicotiana Alata

bs_bs_banner

MOLECULAR PLANT PATHOLOGY (2014) 15(1), 67–79 DOI: 10.1111/mpp.12066

Inhibition of cereal rust fungi by both class I and II defensins derived from the ﬂowers of Nicotiana alata

PETER M. DRACATOS1,2,*, NICOLE L. VAN DER WEERDEN2, KATE T. CARROLL2, ELIZABETH D. JOHNSON2, KIM M. PLUMMER1 AND MARILYN A. ANDERSON2 1Department of Botany, La Trobe University, Melbourne, Vic. 3086, Australia 2La Trobe Institute for Molecular Science, La Trobe University, Melbourne, Vic. 3086, Australia

membranes of Puccinia spp.This study provides evidence that both SUMMARY secreted (NaD2) and nonsecreted (NaD1) defensins may be useful Defensins are a large family of small, cysteine-rich, basic proteins, for broad-spectrum resistance to pathogens. produced by most plants and plant tissues. They have a primary function in defence against fungal disease, although other functions have been described. This study reports the isolation and characterization of a class I secreted defensin (NaD2) from the INTRODUCTION flowers of Nicotiana alata, and compares its antifungal activity with the class II defensin (NaD1) from N. alata flowers, which is The breakdown of crop resistance against phytopathogenic fungi stored in the vacuole. NaD2, like all other class I defensins, lacks is a perennial problem for plant breeders on a global scale. This is the C-terminal pro-peptide (CTPP) characteristic of class II mainly attributed to the planting of crop varieties in genetically defensins. NaD2 is most closely related to Nt-thionin from uniform monocultures, which makes them more susceptible to N. tabacum (96% identical) and shares 81% identity with MtDef4 pathogens that can reduce yield significantly. The economically from alfalfa. The concentration required to inhibit in vitro fungal important pathogens of cereal species include the biotrophs growth by 50% (IC50) was assessed for both NaD1 and NaD2 for [stem rust (Puccinia graminis), leaf rust (P. triticina and P. hordei), the biotrophic basidiomycete fungi Puccinia coronata f. sp. avenae stripe rust (P. striiformis), crown rust (P. coronata), powdery (Pca) and P. sorghi (Ps), the necrotrophic pathogenic ascomycetes mildew (Blumeria graminis), smut (Ustilago tritici) and bunts Fusarium oxysporum f. sp. vasinfectum (Fov), F. graminearum (Tilletia controversa)] and necrotrophs [leaf blotch (Phaeosphaeria (Fgr), Verticillium dahliae (Vd) and Thielaviopsis basicola (Tb), and nodorum) and ear rot (Fusarium graminearum, Fgr)]. Stem rust is the saprobe Aspergillus nidulans. NaD1 was a more potent anti- the most feared disease, especially in wheat and oats, as severe fungal molecule than NaD2 against both the biotrophic and cases have caused total crop failure. The discovery of a new necrotrophic fungal pathogens tested. NaD2 was 5–10 times less pathotype of P. graminis f. sp. tritici in Uganda in 1999 (Ug99)isof effective at killing necrotrophs, but only two-fold less effective on particular concern to breeders and a threat to global food security, Puccinia species. A new procedure for testing antifungal proteins as it carries new virulence to the major resistance (R) gene Sr31, is described in this study which is applicable to pathogens with which renders most of the resistance deployed over the past spores that are not amenable to liquid culture, such as rust patho- 50 years ineffective (Pretorius et al., 2000; Singh et al., 2008). Its gens. Rusts are the most damaging fungal pathogens of many spread throughout the world’s wheat belt, in the Middle East and agronomically important crop species (wheat, barley, oats and southern Asia, has been facilitated by the ability of urediniospores soybean). NaD1 and NaD2 inhibited urediniospore germination, to travel large distances on prevailing trade winds (Ayliffe et al., germ tube growth and germ tube differentiation (appressoria 2008; Hovmøller et al., 2002; Singh et al., 2006). The crown rust induction) of both Puccinia species tested. NaD1 and NaD2 were pathogen of cultivated oats is the most genetically diverse cereal fungicidal on Puccinia species and produced stunted germ tubes pathogen as a result of its co-location and inoculum build up on with a granular cytoplasm. When NaD1 and NaD2 were sprayed wild oats. In Australasia, the crown rust pathogen has developed onto susceptible oat plants prior to the plants being inoculated virulence to nearly every resistance gene deployed in cultivated with crown rust, they reduced the number of pustules per leaf oats over previous decades. Novel host mechanisms of resistance area, as well as the amount of chlorosis induced by infection. are therefore required to provide durable crop protection (Ayliffe Similar to observations in vitro, NaD1 was more effective as an et al., 2008; Lay and Anderson, 2005). antifungal control agent than NaD2. Further investigation Antimicrobial molecules are produced by a diverse range of revealed that both NaD1 and NaD2 permeabilized the plasma organisms, including insects (Bulet et al., 1999), plants (Broekaert et al., 1997; Lay and Anderson, 2005), fungi (Mygind et al., 2005) *Correspondence: Email: [email protected] and mammals (Lehrer and Ganz, 1999), for protection against

© 2013 BSPP AND JOHN WILEY & SONS LTD 67 68 P. M. DRACATOS et al. infection and damage by potential pathogens. Plants produce N. alata (NaD2), and a new bioassay for testing the effect of several different classes of antimicrobial molecules, including antifungal molecules on the early developmental stages of obli- reactive oxygen species, phytoalexins, phytoanticipins and gate biotrophs. The antifungal activity, phenotypic effect and pro- pathogenesis-related proteins (Dangl and Jones, 2001). Recent posed mechanism of activity of NaD1 and NaD2 are reported for research efforts have focused on the transgenic introduction of two different cereal rust species [Puccinia coronata f. sp. avenae antimicrobial proteins from multiple novel sources into agronomi- (Pca) and P. sorghi (Ps)], and are compared with the activity cally important crop species (Girgi et al., 2006; Oldach et al., against a range of necrotrophic fungi. 2001). For example, the incidence of wheat leaf rust (P. triticina f. sp. tritici) and millet rust (P. substriata) was reduced signifi- RESULTS cantly in transgenic wheat and millet plants that expressed afp (antifungal protein from Aspergillus giganteus) within the Isolation and characterization of a type I defensin apoplast, relative to wild-type control plants (Girgi et al., 2006; from N. alata Oldach et al., 2001). Other studies have reported that foliar appli- cations of plant extracts derived from Tulbaghia violacea and The term ‘defensin’ was used as a search query within the sol Agapanthus africanus reduced the pustule number and density of genomics network database (http://www.sgn.cornell.edu) to leaf rust on barley leaves (Cawood et al., 2010; Barna et al., extract representative solanaceous defensin sequences for primer 2008). However, the active constituents responsible for control design and polymerase chain reaction (PCR) amplification of a remain unknown. class I defensin cDNA from Nicotiana alata. Representative class I Plant defensins are small (45–54 residues), cysteine-rich pro- defensin cDNAs derived from Nicotiana tabacum [NTS13 (Li and teins that are produced by many plant species, as well as numer- Gray, 1999) and Nt-thionin (Accession No BAA95697)] were ous plant organs, including leaves, pods, tubers and flowers (Lay selected as candidate sequences and, when aligned, were highly and Anderson, 2005; Thomma et al., 2002). Plant defensins can be conserved at both the nucleotide and amino acid level (90.9% and divided into two major classes. Class I defensins are produced 93%, respectively). Primers designed to these defensins amplified from precursor proteins composed of an endoplasmic reticulum a 234-bp fragment from cDNA derived from N. alata flowers. This (ER) signal sequence and a positively charged defensin domain of fragment encoded a defensin with a predicted ER signal peptide of about 47 amino acids. The ER signal directs them into the ER. They 31 amino acids and a mature defensin domain of 47 amino acids, travel through the secretory pathway and are secreted into the and was named NaD2 (Nicotiana alata Defensin 2) (Fig. 1B). The apoplast. Most of the well-characterized seed defensins belong to predicted protein lacked the CTPP domain which is characteristic this class (van der Weerden and Anderson, 2013). Class II of class II defensins, such as NaD1 (Fig. 1A,B), and shared 96% defensins have an additional C-terminal pro-peptide (CTPP) which and 81% identity with Nt-thionin and NTS13, respectively is negatively charged and is required for vacuolar targeting (Fig. 1C). NaD2 contains the eight highly conserved cysteines that (Fig. 1A). The class II plant defensin, NaD1, is expressed in the define plant defensins, as well as two glycine residues at positions flowers of N. alata and is stored in the vacuole. It functions to 12 and 33, an aromatic residue at position 10 and a glutamate at protect the reproductive organs against attack by necrotrophic position 28, which are present in most, but not all, defensins (van fungal pathogens (Lay et al., 2003; van der Weerden et al., 2008). der Weerden and Anderson, 2013). The mass predicted from the NaD1 binds to the hyphal cell wall of F. oxysporum f. sp. cDNA clone (5255.9 Da) was the same as that of the NaD2 protein vasinfectum (Fov), where it interacts with either the glycoprotein purified from flowers (5254.2 Da). or β-glycan layer (van der Weerden et al., 2008, 2010). After NaD2 falls into a group of class I defensins that is widely passing through the cell wall, NaD1 exerts its antifungal activity by represented throughout the plant kingdom and includes defensins permeabilizing the plasma membrane and entering the cytoplasm from lucerne (Medicago truncatula, MtDef4) (Sagaram et al., of Fov hyphae (van der Weerden et al., 2008). NaD1 forms dimers 2011), cowpea (Vigna unguiculata, JI-2) (Urdangarin et al., 2000) in solution which are essential for its antifungal activity (Lay et al., and African oil palm (Elaeis guineensis, EGAD1) (Tregear et al., 2012). 2002) (Fig. 1C). NaD2 was 74%–96% identical to defensins from NaD1 has potent in vitro antifungal activity against filamentous this group, but only 32%–42% identical to solanaceous class II ascomycete fungi. The concentration of NaD1 required to inhibit defensins, and less than 30% identical to more distantly related fungal growth by 50% (IC50) is as low as 0.8 μM for some species class I defensins from radish and dahlia (Fig. 1C). (van der Weerden et al., 2008). The ability of NaD1 and secreted defensins to inhibit the growth of obligate biotrophs, which Expression of NaD2 in N. alata flowers include some of the most agronomically significant basidiomycete pathogens, has not been reported. Here, we report the isolation NaD2 and NaD1 were only expressed in the flowers of N. alata and characterization of a class I defensin from the flowers of and were not detected in stems, leaves or roots (Fig. 2). NaD2, like

Fig. 1 (A) Diagram of precursors encoding class I and class II defensins. Both precursors have endoplasmic reticulum (ER) signal sequences that direct the protein into the secretory pathway. Class I defensins have no additional targeting information and are secreted. Class II defensins have a C-terminal pro-peptide (CTPP). In the solonaceous class II defensins, this pro-peptide targets the defensin to the vacuole, where the CTPP is removed to release the mature domain. Processing sites are indicated by black arrows. (B) Full-length amino acid sequence alignment of NaD2 with the class I defensins Nt-thionin (Accession No AB034956) and NTS13 (Accession No X99403) (Nicotiana tabacum) and the class II defensin NaD1 (N. alata) (van der Weerden and Anderson, 2013). ER signals are indicated in italics and the C-terminal prodomain of NaD1 is underlined. Processing sites are indicated by an arrow. (C) Amino acid alignment of NaD2 with mature class I defensin sequences from N. tabacum (Nt-thionin and NTS13), Capsicum annum (CaDef2), Medicago truncatula (MtDef4), Helianthus annuus (JI-2), Elaeis guineensis (EGAD1), Vigna unguiculata (Cpthio1), Raphanus sativus (RsAFP2) and Dahliae merckii (DmAMP1), and mature class II defensin sequences from N. alata (NaD1), N. tabacum (FST), Petunia hybrida (PhD1) and Solanum lycopersicum (TPP3). Disulphide bonds between the eight conserved cysteines are shown by connecting lines. Loops (L1–L7) are classiﬁed as the regions between cysteine residues (van der Weerden and Anderson, 2013).

NaD1, was most abundant in the sepals, but NaD2 was less Antifungal activity of N. alata defensins prominent within the ovaries relative to NaD1 (Table 1). NaD2 was most abundant at stage I of floral development (initial bud stage) NaD1 and NaD2 were purified from flowers and tested for their and gradually declined as a percentage of total protein as the in vitro (in liquid growth media) antifungal activity against a flower matured (Table 1). NaD2 was less abundant than NaD1 at broad range of filamentous ascomycete pathogens including Fov, all stages of floral maturity and within all different floral organs Fgr, Verticillium dahliae (Vd), Thielaviopsis basicola (Tb) and the tested, as measured using enzyme-linked immunosorbent analysis saprobe Aspergillus nidulans (An) (Table 2). NaD1 was signifi-

(ELISA) on total protein extracts from N. alata ﬂowers (Table 1). cantly more effective than NaD2 (4–14 times higher IC50) at killing

IC50 values for NaD1 and NaD2 against both Puccinia rust pathogens are shown in Table 2. Urediniospores which germinated in the treated sample were stunted and did not resemble normal germ tubes, as observed in the controls (Fig. 3A). NaD1 was more effective against Ps than Pca at the high concentrations tested,

especially at 5 μM (Fig. 3B). NaD2 had higher IC50 values than NaD1, but still inhibited germination signiﬁcantly relative to the negative control protein, NaPI (Heath et al., 1997) (Fig. 3B). NaPI had no effect on urediniospore germination and spores appeared to have the same germination frequency as the double-distilled

H2O (ddH2O) control (Fig. 3B).

Effect of N. alata defensins on germ tube growth and appressorium formation

Both NaD1 (Figs 4 and 5) and NaD2 (Fig. 5) arrested germ tube elongation when they were added to Pca and Ps germlings, 80 min after germination. Growth arrest from NaD1 treatment was Fig. 2 Expression of defensins in Nicotiana alata. Proteins extracted from flowers (F), stems (S), leaves (L) and roots (R) and separated by sodium assessed 4 h after the addition of defensin. Growth arrest occurred dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Defensins in 70% of total scored germlings for both Ps and Pca (Fig. 4B).The were detected by immunoblotting with antibodies raised against NaD2 (A) defensins appeared to be fungicidal as the growth arrest of or NaD1 (B). M, SeeBlue® plus two prestained protein standard (Life Puccinia germ tubes was persistent 24 h after application of either Technologies) size markers. The NaD2 and NaD1 defensins were only NaD1 or NaD2. NaD1 treatment resulted in the stunting of germ detected in flowers. Recombinant defensins (rNaD1 and rNaD2) were used as tubes and a granular appearance of the cytoplasm in spores and positive controls. germ tubes of both Pca and Ps (Fig. 4A). NaD2 produced the same all of these pathogens (nonbiotrophic ascomycetes) (Table 2). effect (Fig. 5). Four hours after spore germination, the germ tubes of were The IC50 for NaD1 was less than or equal to 1 μM for all five Pca ascomycetes. In contrast, NaD2 was most active against Fgr subjected to a heat treatment for 2 h (30 °C) in the presence of 0.5 mM -2-hexen-1-ol in ddH O to induce appressoria forma- (IC50 =2μM) and was less active against Fov, Tb and An (IC50 = trans 2 5–7 μM). NaD2 did not inhibit the growth of Vd at the concentration on fresh agar plugs. When NaD1 was applied to plugs that tions tested. had been treated for appressorial induction 4 h after spore germination, a significant reduction (66.6%) in the number of Development of a bioassay to test the effects of appressoria was observed relative to the water control (Fig. 4C). N. alata defensins on rust fungi Permeabilization of rust germ tubes by An in vitro bioassay was developed to assess the effect of anti- N. alata defensins fungal proteins on spore germination, germ tube growth and differentiation of pathogens with hydrophobic spores (not ame- The extent to which NaD1 and NaD2 permeabilized the plasma nable to liquid culture). Wetting agents were trialled; however, membrane of Ps germ tubes was measured using the fluorescent these interfered with the bioassay (data not shown). The bioassay nucleic acid dye, SYTOX green (Life Technologies, Grand Island, NY, described here is based on the inoculation of urediniospores onto USA) (Fig. 5). SYTOX green dye cannot pass through an intact 1% water agar, followed by the excision of 7-mm-diameter plugs plasma membrane and only binds to nucleic acids of cells with of equal volume, which were arrayed onto microscope slides. In all compromised membranes. SYTOX green fluorescence was detected cases, fresh urediniospores from living plants exhibited optimal 15 min after application of the defensins to the germ tubes of both viability, that is, had a germination frequency of >85%. Puccinia species, but was not detected in the urediniospore (Fig. 5) (data not shown for Pca). Antifungal effect of N. alata defensins on urediniospore germination Effects of NaD1 and NaD2 as foliar agents against oat crown rust NaD1 and NaD2 inhibited the germination of both of the Puccinia rust pathogens tested (Fig. 3A,B). The most dramatic inhibition The foliar application (200 μg/mL) of both NaD1 and NaD2 on was for Pca urediniospores treated with 10 μM NaD1 (Fig. 3A).The crown rust-susceptible oat plants resulted in reduced pustule

Table 1 Percentage of NaD2 and NaD1 in total protein extracts derived from different stages of development and floral organs from Nicotiana alata flowers. Total soluble protein extracted from N. alata flowers was analysed by enzyme-linked immunosorbent analysis (ELISA). The relative level of defensins to total soluble protein decreased as the flowers reached maturity.

Floral developmental Nicotiana alata stage NaD2 (%) NaD1 (%) ﬂoral organ NaD2 (%) NaD1 (%)

I 0.35 2.3 Petals 0.04 0.14 II 0.26 1.8 Sepals 0.08 0.38 III 0.28 1.6 Anthers 0.01 0.13 IV 0.17 1.6 Styles 0.02 0.13 V 0.08 0.4 Stigmas 0.02 0.05 Ovaries 0.02 0.23

Table 2 The concentrations of NaD1 and NaD2 IC50 (μM) required to reduce fungal growth by 50% (IC50) of the necrotrophic fungal species Aspergillus Microbial classification Organism tested NaD1 NaD2 IC50NaD2/IC50NaD1 nidulans and the obligate biotrophic rust species Puccinia coronata f. sp. avenae and Fungi Ascomycota f. sp. 155 Puccinia sorghi.IC values (NaD1) against Fusarium oxysporum vasinfectum 50 Fusarium graminearum 0.5 2 4 fungal species within the Ascomycota were Verticillium dahlia 0.75 >10 >13.5 sourced from van der Weerden et al. (2008). Thielaviopsis basicola 177 Aspergillus nidulans 0.8 5 6.3 Basidiomycota Puccinia coronata f. sp avenae 2.5 4 1.6 Puccinia sorghi 2 5 2.5 frequency and increased photosynthetic area relative to the oat NaD2 falls into a subgroup of class I defensins that is widely leaves sprayed with water prior to inoculation (Fig. 6).As observed represented throughout the plant kingdom. Members of this during antifungal activity assays, NaD1 was more effective than group can be found in Arabidopsis, sunflower, alfalfa and the NaD2 at reducing the pustule number per leaf area (70% versus African oil palm, as well as other solanaceous species, such as 36%) and maintaining the photosynthetic area, relative to control capsicum and petunia (van der Weerden and Anderson, 2013). plants that were severely infected and chlorotic at 10 days post- Several of these defensins have antifungal activity (Tregear et al., infection (Fig. 6). 2002; Urdangarin et al., 2000), although the mechanism of antifungal activity has only been studied in detail for one of these, MtDef4. MtDef4 permeabilizes the membrane of target DISCUSSION hyphae, but does not require the presence of the sphingolipid Plant defensins have been isolated from a number of diverse glucosylceramide (GlcCer) for activity (Ramamoorthy et al., 2007). ornamental and crop species, and have been tested for their The residues in loop five, which form part of the γ-core motif, have antifungal properties on several agronomically significant fungal been implicated in the antifungal activity of plant defensins (Yount pathogens (for a review, see van der Weerden and Anderson, and Yeaman, 2004). Residues in this loop have been reported to be 2013). The potent antifungal activity of NaD1, a class II defensin crucial for the antifungal activity of RsAFP2, MsDef1 and MtDef4 from N. alata flowers, has been reported against numerous fila- (De Samblanx et al., 1997; Sagaram et al., 2011; Schaaper et al., mentous ascomycete fungi, including Fov, Fgr, Leptosphaeria 2001; Spelbrink et al., 2004). The γ-core motif of MtDef4 has been maculans and Botrytis cinerea (Lay et al., 2003; van der Weerden demonstrated to be largely responsible for antifungal activity. et al., 2008).The class I defensin NaD2, characterized in this study, When Sagaram and co-workers grafted the γ-core motif of MtDef4 is also produced by the flowers of N. alata, but is a weaker onto MsDef1 (a similar class I defensin with weaker antifungal (2–10 times) antifungal molecule than NaD1, especially on activity), the hybrid molecule showed antifungal activity against necrotrophic fungal pathogens. The IC50 values obtained for NaD2 Fgr that was equivalent to MtDef4 (Sagaram et al., 2011). Further- on necrotrophic ascomycete pathogens were similar to those more, the hybrid molecule retained activity against an MsDef1- of other well-characterized class I plant defensins, such as resistant mutant of Fgr that lacked GlcCer (Sagaram et al., 2011). DmAMP1 (Dahlia merckii) (Thevissen et al., 2000) and MtDef4 NaD2 and MtDef4 share a high degree of similarity and may (M. truncatula) (Ramamoorthy et al., 2007), whereas NaD1 therefore share a similar mode of action. Based on the sequence showed similar inhibitory activity to that reported for the radish similarity between MtDef4 and NaD2, it is also likely that the defensin RsAFP2 (Terras et al., 1992). γ-core motif is involved in the antifungal activity of NaD2.

Fig. 3 The inhibitory effects of defensin on urediniospore germination. (A) Micrographs of

the effect of double-distilled water (ddH2O) (i) on uredinospore germination of Puccinia coronata f. sp avenae (Pca) compared with treatment with 10 μM NaD1 (ii). Scale bar, 20 μm. (B) Graph representation of the effect of NaD1 and NaD2 on urediniospore germination of two rust pathogens Pca and Puccinia sorghi (Ps). Full and broken lines represent Pca and Ps, respectively, and squares, triangles and diamonds represent assays using NaD1, NaD2 and NaPI protein treatments, respectively. All experiments were repeated three times and at least 100 urediniospores were scored on each agar plug. (C) Graphic representation of germ tube growth (μm) on 1% water agar plates versus time (min) after germination. In all cases, error bars represent the standard error of the mean of four biological replicates.

A bioassay on artiﬁcial media was developed to assess the Potter et al., 1990). This may explain the observation that some effect of defensins on different developmental stages of obligate urediniospores germinated faster than others, or appeared to be biotrophs, such as the cereal-infecting Puccinia rusts.As a result of less susceptible to defensin treatment. NaD1 inhibited rust germi- obligate biotrophy and the hydrophobicity of rust urediniospores, nation at comparatively higher concentrations (IC50 < 2.5 μM) the defensins could not be assessed using aqueous assays per- than the ascomycete fungi (IC50 < 1 μM), such as Fov, Fgr and formed in microtitre plates. Water agar plugs mounted on micro- L. maculans (Lay et al., 2003; van der Weerden et al., 2008). scope slides supported urediniospore hydration and efﬁcient Higher IC50 inhibitory concentrations observed for the Puccinia application of the defensins. Field-derived isolates of rust spores species tested in this study may be a result of absorption and are often not only genetically diverse, but can differ in virulence to subsequent dilution of NaD1 within the water agar matrix. their hosts (Anikster and Wahl, 1979; Burdon, 1987; Dracatos Both defensins inhibited the germination of urediniospores and et al., 2008, 2009; McDonald and Linde, 2002; Park et al., 2011; elongation of germ tubes,and were toxic against both Puccinia spp.

Fig. 4 (A) Micrograph images of 10 μM NaD1-treated germlings of Puccinia coronata f. sp avenae (Pca) (iii), Puccinia sorghi (Ps) (iv) and their corresponding water controls (i) and (ii). Arrows denote examples of stunted germlings with granular cytoplasm. Scale bar, 20 μm. (B) The effect of 10 μM NaD1 treatment on germ tube growth from germinated Pca and Ps (white) urediniospores relative to untreated controls (black). Germlings were classiﬁed as stunted if, 4 h after defensin treatment, their length was less than twice the diameter of the urediniospore. The percentage of stunted germ tubes was calculated as the ratio of stunted to total germ tubes multipled by 100. In all cases, error bars represent the standard error of the mean of four replicates. All experiments were repeated three times and at least 100 urediniospores were scored on each agar plug. (C) The effect of NaD1 treatment, applied 4 h after urediniospore germination, on the differentiation of Pca germlings, as measured by appressorium production on artiﬁcial media. Induction of appressoria was performed by treatment with 0.5 mM trans-2-hexen-1-ol together with heating at 30 °C for 2 h. Appressorium production was the most reproducible measurement of differentiation for experimental purposes using this method.

tested. Further evidence of the fungicidal activity of the defensins Defensins are encoded by large multi-gene families (300 in includes the observations of tip swelling and granulation of the Arabidopsis thaliana and Medicago spp.) (Silverstein et al., 2005, cytoplasm in the germ tubes of Ps and Pca, respectively. In this 2006). It is likely that different defensins have evolved in plant study, both NaD1 and NaD2 permeabilized the fungal plasma species to provide protection against pathogens with different membrane of the Puccinia spp. tested. NaD1 and NaD2 were also mechanisms of invasion and with diverse cell wall composition. able to protect oat seedlings from heavy infection by Pca spores Both NaD1 and NaD2 were able to inhibit the in vitro germination when the defensins were applied to the foliage as a spray. and germ tube growth of Ps and Pca, and NaD1 inhibited germ

Fig. 5 Micrographs of Puccinia sorghi (Ps) germlings treated with SYTOX green dye and

double-distilled water (ddH2O) (i and ii), 10 μM NaD1 (iii and iv) and 10 μM NaD2 (v and vi). Note that fluorescence caused by SYTOX green uptake is only visible in Ps germ tubes that have been exposed to both SYTOX green and NaD1 or NaD2 treatment. Images were viewed under white light (i, iii and v) and under a 460–490-nm MWIB filter (ii, iv and vi). Scale bar, 20 μm. tube differentiation. Although NaD2 had lower antifungal activity transgenic plant was produced with enhanced fungal resistance than NaD1 on necrotrophic fungi (4–14 times less active), the (Terras et al., 1995), it is interesting that none have been commer- difference was less marked against Puccinia species, where it was cialized to date. One example of a class I defensin that showed two-fold less active than NaD1. Based on the presence of an ER particular promise was alfAFP (also known as MsDef1) from alfafa signal peptide, NaD2 is likely to be secreted into the apoplast (Gao et al., 2000). Constitutive expression of alfAFP in potatoes and would therefore be more likely to come into contact with conferred enhanced resistance (approximately six-fold reduction biotrophic pathogens, such as rust fungi, than would NaD1, which in fungal load in the transgenic versus nontransgenic plants) to is targeted to the vacuole. A recent study by Kaur et al. (2012) early dying disease caused by Vd. Furthermore, the protection demonstrated that defensins targeted to the apoplast, but not to conferred by alfAFP was maintained under both glasshouse and intracellular compartments, can provide protection against the field conditions over several years at different geographical sites obligate biotroph Hyaloperonospora arabidopsidis. As a conse- (Gao et al., 2000). However, plants expressing alfAFP also showed quence, NaD2 could be well suited for in planta expression. The a reduction in potato tuber size (Seale and Vordtriede, 2010). stunting of germ tubes on exposure to defensin, reported by Kaur Perhaps this and other unreported undesirable agronomic effects et al. (2012), was similar to that observed in this study. of transgene expression may have limited the commercial appli- Almost 20 unique plant defensins have been transformed into cation of plant defensins for disease control in crop plants. various plants (for examples, see review by De Coninck et al., Recently, a class II defensin from petunia flowers was used to 2013). Given the relative success of producing transgenic plants produce transgenic banana resistant to Fusarium oxysporum f. sp. with defensins and the lapse of time since the first defensin Cubense (Ghag et al., 2012). These banana plants were pheno-

In this study, we report the isolation and characterization of a type I defensin from N. alata (NaD2) and comparison of the antifungal activity of both NaD1 and NaD2 on necrotrophic and biotrophic pathogens.We have determined that defensins are able to kill rust urediniospores and germ tubes, and prevent the differentiation of rust germlings. Foliar application of defensins also reduces the infection of oat plants by Pca. Further research is required to determine whether transgenic crop plants expressing either NaD1 or NaD2 will provide resistance to rust fungi in the ﬁeld.

EXPERIMENTAL PROCEDURES

Cloning and PCR ampliﬁcation of NaD2

Total RNA was extracted from immature stigmas (pre-anthesis) of N. alata flowers using TRIzol®, essentially as described by the manufacturer (Life Technologies). The RNA was precipitated with diethyl-polycarbonate (DEPC)-treated lithium chloride (final concentration of 2 M), collected by centrifugation and washed with 70% ethanol that had been pre-treated with DEPC. The quality of the RNA was assessed by gel electrophoresis (1% agarose). cDNA was synthesized from total RNA using a Superscript® III First Strand Synthesis Kit (Life Technologies). Any remaining RNA was removed by incubation with 2 U RNase H (Life Technologies) at 37 °C for 20 min. Forward (GGATCCATGGCAAACTCCATGCCG) and reverse (GAGCTC TTAGCAAGGCCTGGTACAGA) primers were designed on the basis of the Nt-thionin cDNA sequence (Accession No AB034956) to amplify the coding sequence of the N. alata homologue of Nt-thionin (ER signal sequence and mature domain) and included both BamHI and SacI restriction sites at the 5' ends (italics). The NaD2 cDNA was amplified in a 50-μL PCR that contained 100 ng of stigma cDNA, Thermapol buffer (New England Biolabs, Ipswich, MA, USA), 0.2 mM deoxynucleoside triphosphate (dNTP) mix (Promega, Madison, WI, USA), 2.5 U Taq DNA polymerase (New England Biolabs) and 0.4 μM of forward and reverse primers. The PCR cycling conditions were as follows: 94 °C for 5 min for the initial denaturation step; 30 cycles of 94 °C for 1 min, 63 °C for 1 min, 72 °C for 1 min; and a final extension period of 72 °C for 10 min. PCR products were separated using gel electrophoresis and purified using the Fig. 6 Images of whole oat plants (A) and oat leaves (B) sprayed with NaD2 Wizard® SV clean-up system (Promega), cloned into the pCR® 2.1-TOPO® (left and top), NaD1 (centre) or water (right and bottom) prior to inoculation vector (Life Technologies) and transformed into Escherichia coli TOP10 with Puccinia coronata spores. (C) Percentage reduction in pustules per cells (Life Technologies), according to the manufacturer’s instructions. square centimetre of infected leaves for NaD2-treated (black) and Plasmid DNA was extracted using the QIAprep® Spin Miniprep kit (Qiagen, NaD1-treated (white) plants relative to water-treated leaves (control) (grey). Venlo, Netherlands) and positive clones were confirmed by sequence NaD1 and NaD2 reduced the number of pustules per square centimetre by analysis using both M13 forward and reverse primers. 70% and 37%, respectively. typically normal; however, banana plants expressing the petunia Isolation of floral defensins defensin mature domain in the absence of the CTPP were severely NaD1 and NaD2 were extracted from flowers essentially as described in stunted (Ghag et al., 2012). This indicates that vacuolar targeted van der Weerden et al. (2008). The proteins were eluted as distinct peaks class II defensins may have less phytotoxic effects than their class on reverse-phase high-performance liquid chromatography (RP-HPLC), I counterparts, making them a superior choice for the production and the identity and purity of each protein were confirmed by mass of commercially viable transgenic crops. spectroscopy.

were lyophilized on a DYNAVAC FD5 freeze drier. The antibodies were Expression of recombinant NaD1 and NaD2 (rNaD1 and dissolved in 1 mL PBS, pH 7.5, and incubated on ice for 2 h at a molar rNaD2) in Pichia pastoris ratio of 1:20 with sulfo-NHS-LC-Biotin (Pierce). The concentration of Expression of rNaD1 and rNaD2 was performed in the methylotrophic biotin-labelled antibody was estimated by absorbance at 280 nm. The α α yeast Pichia pastoris. A DNA fragment encoding the mature NaD1 -NaD2 (100 ng per well) and -NaD1 (50 ng per well) antibodies were protein was amplified by PCR using the following forward and reverse absorbed to the base of the wells of a 96-well microtitre plate by incu- primers: forward (CTCGAGAAAAGAGCTAGAGAATG), which added the bation at 4 °C in a humid box overnight. The wells were then washed × XhoI restriction endonuclease site (bold) and the KEX2 cleavage site with PBS (4 2 min) and blocked with 3% (w/v) bovine serum albumin (italics); reverse (GCGGCCGCTTACATGGCTTAGTAC), which added a NotI (BSA) (Sigma-Aldrich, St. Louis, MO, USA; ELISA grade) in PBS for 2 h at restriction endonuclease site (bold) and a stop codon (italics). Primers room temperature, and then rewashed as described above. Protein were designed to remove the STE13 protease site between the KEX2 extracts were prepared by grinding 100 mg of liquid nitrogen-frozen × cleavage site and NaD1 to prevent the addition of glutamic acid–alanine tissue in a mixer mill (2 10 s, frequency 30) before the addition of μ (Glu–Ala) repeats at the N-terminus of NaD1. An Ala was added to the 900 L of 2% insoluble polyvinylpyrrolidone (PVPP) (Sigma-Aldrich) in μ N-terminus to ensure efficient cleavage by KEX2. A similar approach was PBST [PBS; 0.05% (v/v) Tween-20]. Protein extracts (100 L) were added used for NaD2 with the following forward and reverse primers: forward to the wells and the plates were incubated for 2 h at 25 °C. NaD1 and μ μ CTCGAGAAAAGAGCTAGAACTTGCGAG (XhoI restriction site in bold, NaD2 protein standards (100 L, 0.1–0.0015 ng/ L) were prepared in KEX2 cleavage site in italics) and reverse GCGGCCGCTTAGCA PBS and added to the wells at the same time as the protein extracts. × AGGCCTGGT (NotI site in bold, stop codon in italics). Further cloning Excess extract was removed by washing with PBS (4 2 min) prior to α α and transformation were performed essentially as described by the the addition of -NaD2 (150 ng per well) or -NaD1 (50 ng per well) manufacturer (Life Technologies). Electrocompetent Pichia pastoris biotin-labelled antibodies in PBS. The plates were then incubated for 1 h GS115 cells (Life Technologies) were prepared as described by Chang at 25 °C. Excess antibody was removed by again washing with PBST × et al. (2005), and linearized DNA was transformed into these cells using (4 2 min) prior to the addition of NeutriAvidin horseradish peroxidase μ μ standard protocols. NaD1 and NaD2 were then expressed in buffered (HRP)-conjugate (0.1 L/well in 100 L PBST, Thermo Fisher Scientific). minimal media according to the manufacturer’s instructions and purified Plates were incubated for 1 h at 25 °C before washing with PBST × × via cation-exchange chromatography. Eluted proteins were subjected to (4 2 min) and water (2 2 min). The ImmunoPure OPD substrate RP-HPLC using a 40-min linear gradient as described by Lay et al. (Thermo Fisher Scientific) was prepared according to the manufacturer’s μ (2003). Protein peaks were collected and analysed by sodium instructions, and 100 L of substrate solution were added to each well. dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and Plates were allowed to develop for 5 min before the reaction was μ immunoblotting with anti-NaD1 and anti-NaD2 antibodies (see below). stopped by the addition of 50 Lof2.5M sulphuric acid. Absorbance at Fractions containing the protein of interest were lyophilized and 490 nm was measured using a SpectraMax M5e microtitre plate reader resuspended in sterile MilliQ ultrapure water (Merck Millipore, Billerica, (Molecular Devices, Sunnyvale, CA, USA). MA, USA). The protein concentration was determined using the bicinchoninic acid (BCA) protein assay (Thermo Fisher Scientific, Rockford, IL, USA). Fungal isolates Field isolates (mixture of genotypes) of oat crown rust (Pca) were provided by Dr Philip Keane (La Trobe University, Melbourne, Vic., Australia) Production of anti-NaD1 and anti-NaD2 antisera and common maize leaf rust (Ps) urediniospores were provided by Polyclonal antibodies were raised against purified NaD1 or NaD2 (1.5 mg) Dr Tony Pryor (CSIRO Plant Industries, Canberra, Australia). All bio- that had been conjugated to keyhole limpet haemocyanin (0.5 mg) and assays were performed, where possible, with freshly harvested injected into a rabbit, essentially as described by Lay et al. (2003). urediniospores. Five other fungal isolates were assessed for sensitivity to both NaD1 and NaD2: Fgr (Australian isolate CS3005; provided by CSIRO Plant SDS-PAGE and immunoblot analysis Industry, St Lucia, Qld, Australia); Fov (Australian isolate VCG01111 isolated from cotton; provided by Wayne O’Neill, Farming Systems Institute, Soluble protein (30 μg) from whole flowers and from all other tissues DPI, Qld, Australia); Vd (provided by Helen McFadden, CSIRO Plant Indus- (200 mg) was analysed for expression using both α-NaD1 and α-NaD2 try, Black Mountain, Australia); Tb (provided by Dr David Nehl, NSW antibodies, as described by Lay et al., 2003. Department of Primary Industry, Narrabri, Australia); An (strain A850; provided by Professor Michelle Momany, University of Georgia, Athens, GA, USA). Quantification of NaD1 and NaD2 levels using ELISA

α α Prior to use in an ELISA, both the -NaD1 and -NaD2 antibodies were Antifungal bioassays on ascomycetes biotinylated. Protein A-puriﬁed antibodies (2 mg) were passed through a PD-10 column (GE healthcare Bio-Sciences, Uppsala, Sweden) to Antifungal bioassays were prepared essentially as described by van der exchange the glycine with phosphate-buffered saline (PBS) before they Weerden et al. (2008).

Production of rust spores for in vitro bioassays Measurement of the effect of NaD1 on crown rust appressoria induced on artificial media Rust spores were prepared essentially as described by Dracatos et al. (2010). Separate spore suspensions were prepared for Pca and Ps depend- Appressoria were induced on 1% water agar plugs by a modified method ing on the host to be inoculated. Briefly, urediniospore suspensions were previously described by Wiethölter et al. (2003). This method is based on prepared containing 5 × 106 spores/mL in 25 mL of a stock solution (one the presence of a humid environment in combination with both physical drop of Tween-20 in 100 mL of tap water) and decanted into a 500-mL (heat shock) and chemical (trans-2-hexen-1-ol) (Sigma-Aldrich) stimuli. conical glass flask attached to a Venturi atomizer. Spore suspensions were Crown rust urediniospores (2 mg) were spread evenly onto a Petri dish mixed vigorously and the atomizer was used to spray inoculate susceptible (containing 10 mL of 1% water agar) using a fine paintbrush, and six plugs 6-week-old oat plants (Avena sativa cv. Swan, provided by Dr Philip Keane were removed as described previously and placed on the lid of a 5-cm Petri La Trobe University) or maize seedlings (Pioneer 33V15, provided by Dr Joe dish (repeated for four Petri dishes to give a total of 24 plugs). Spores were Kochman, Department of Primary Industry, Toowoomba, Qld, Australia). left to germinate for 90 min at 20 °C, and then 5 mL of 0.5 mM trans-2- Plants were inoculated at 17.00 h, followed by an incubation period of hexen-1-ol were added to the bases of the Petri dishes, and the lids with 16 h in the dark at 20 °C within a dew chamber, before plants were placed plugs were placed on the Petri dish bases and sealed with Parafilm, in the glasshouse at 25 °C. followed by heat treatment at 30 °C for 2 h. NaD1 (10 μLof10μM)was applied 4 h post-germination and water was applied as a control.All steps were performed in the dark. Appressoria numbers were counted (as a In vitro bioassay to determine the effect of NaD1 and percentage of germinated spores) after 24 h by examining eight individual NaD2 on Puccinia urediniospore germination and germ plugs per treatment under an Olympus BX50 microscope (Olympus, Tokyo, tube growth Japan). Both NaD1 and NaD2 that had been isolated from N. alata flowers were assessed for their inhibitory effect on urediniospore germination (Pca and Ps).The germ tube growth of Pca and Ps, after exposure to NaD1, was also Membrane permeabilization assay assessed, as was the ability of the germ tubes to differentiate and produce Ps and Pca urediniospores were brushed onto water agar and incubated at appressoria (Pca) in the presence of NaD1. A 6-kDa serine proteinase 20 °C for 1 h in the dark. Agar plugs with germinated spores, with germ inhibitor protein from N. alata (NaPI) with no reported antifungal activity tubes of approximately 60 μm in length, were treated in duplicate with (Heath et al., 1997) was used as a negative control. Preparation of the either 10 μLof10μM NaD1 or NaD2 containing 0.5 μM SYTOX green in bioassay involved the collection of fresh urediniospores from the surface dimethylsulphoxide (DMSO) (Molecular Probes), which was applied in a of pustules on oat leaves with a fine wet paintbrush, and spreading of the darkened room and left to stand for 10 min. In all cases, a distilled water urediniospores onto 9-cm-diameter Petri dishes containing 10 mL of 1% control containing 0.5 μM SYTOX green and no defensin and a second water agar. An inverted 1-mL sterile pipette tip was used to extract water control with 10 μM NaD1 or NaD2 and no SYTOX green were included in agar plugs with the inoculum (diameter, 7 mm) from the centre of the the bioassay. Images were captured with an Olympus BX50 fluorescence plate. Plugs were placed on a microscope slide using a sterile scalpel. microscope using an excitation wavelength of 460–490 nm and an MWIB Urediniospores were left to hydrate for 20 and 75 min in the dark at 20 °C filter. for germination and germ tube growth assays, respectively. For the assessment of urediniospore germination, plugs treated with a μ total of six different protein concentrations (10, 5, 2, 1, 0.5 and 0.1 M), Measurement of the effect of NaD1 and NaD2 on as well as water as a nonprotein-treated control, were included on each crown rust infection of oat plants slide (10 μL per plug). Four replicate slides were used per experiment and each experiment was repeated three times. Plates were then left to incu- Oat seeds (cv. Swan) were sown in 9-cm pots (14 seeds/pot) filled with a bate in the dark at 20 °C for 150 min (when the inoculated control plate 2:1 mixture of pine bark and river sand.An initial dose of the water-soluble started to produce germ tubes that were >80 μm in length) (Fig. 3C). The fertilizer Aquasol (10 g/10 L of tap water) was applied before sowing.After germination of urediniospores was measured by counting spores versus 7 days, both Aquasol and urea were applied (10 g/10 L of tap water). germlings (spores with germ tubes >40 μm in length) under a light micro- Lyophilized defensin (1.2 mg) was resuspended in sterile MilliQ water to scope under 100× magnification. Four experimental replicates were used 1 mg/mL, diluted to 200 μg/mL and then 3 mL of each treatment (NaD1, and two readings were taken per replicate from the centre of the inocu- NaD2 and water) was sprayed separately onto 8-day-old seedlings (eight lated water agar plugs. In all cases, at least 100 urediniospores and/or seedlings per pot × four seedling pots per treatment) using a hydrocarbon germ tubes were scored for each treatment replicate. The mean germina- pressure pack. Forty five minutes later, the seedlings were inoculated with tion percentage (%) was calculated for each treatment and plotted a pathotype of Pca that is virulent on Swan by the spraying (5 × 106 against the protein concentration (μM). The defensin concentration that spores/mL) of urediniospores suspended in light mineral oil (Isopar-L, inhibited growth by 50% (IC50) was then determined from germination ExxonMobil Chemical, Houston, TX, USA) using a hydrocarbon pressure inhibition plots. Rust germ tubes were scored as stunted if their length pack. Inoculated plants were incubated overnight in a dew chamber at was less than twice the diameter of the spore and nonstunted when their 20 °C in the dark, before being transferred to a growth cabinet (22–23 °C). appearance was the same as that of the untreated control (e.g. >40 μm Plants were subjected to ambient natural light regimes. Representative in length). photographs of each treatment were taken at 10 days post-infection (both

whole plants and leaf sections), and the mean pustule number per square Heath, R., McDonald, G., Bateman, K., Christeller, J., Lee, M., van Heeswijck, R. centimetre of leaf area was calculated on the basis of eight measurements and Anderson, M. (1997) Proteinase inhibitors from Nicotiana alata enhance plant resistance to insect pests. J. Insect Physiol. 9, 833–842. (two leaves per pot) per treatment. Hovmøller, M.S., Justesen, A.F. and Brown J.K.M. (2002) Clonality and long-distance migration of Puccinia striiformis f.sp. tritici in north-west Europe. Plant Pathol. 51, 24–32. ACKNOWLEDGEMENTS Kaur, J., Thokala, M., Robert-Seilaniantz, A., Zhao, P., Peyret, H., Berg, H., Pandey, S., Jones, J. and Shah, D. (2012) Subcellular targeting of an evolutionarily conserved We thank Dr Phillip Keane for assistance with microscopy and Dr Tony plant defensin MtDef4.2 determines the outcome of plant–pathogen interaction in Pryor and Professor Robert Park for the provision of rust isolates used in transgenic Arabidopsis. Mol. Plant Pathol. 13, 1032–1046. Lay, F. and Anderson, M. (2005) Defensins—components of the innate immune this study. system in plants. Curr. Protein Pept. Sci. 6, 85–101. Lay, F., Mills, G., Poon, I., Cowieson, N., Kirby, N., Baxter, A., van der Weerden, N., Dogovski, C., Perugini, M., Anderson, M., Kvansakul, M. and Hulett, M. (2012) REFERENCES Dimerisation of plant defensin NaD1 enhances its antifungal activity. J. Biol. Chem. 287, 19 961–19 972. Anikster, Y. and Wahl, I. (1979) Coevolution of the rust fungi on Gramineae and Lay, F.T., Brugliera, F. and Anderson, M.A. (2003) Isolation and properties of floral Liliaceae and their hosts. Annu. Rev. Phytopathol. 17, 367–403. defensins from ornamental tobacco and petunia. Plant Physiol. 131, 1283–1293. Ayliffe, M., Singh, R. and Lagudah, E. (2008) Durable resistance to wheat stem rust Lehrer, R. and Ganz, T. (1999) Antimicrobial peptides in mammalian and insect host needed. Curr. Opin. Plant Biol. 11, 187–192. defence. Curr. Opin. Immunol. 11, 23–27. Barna, B., Leiter, É., Hegedus, N., Bíró, T. and Pócsi, I. (2008) Effect of the Penicillium Li, H.Y. and Gray, J.E. (1999) Molecular characterization of a cDNA, NTS13, encoding chrysogenum antifungal protein (PAF) on barley powdery mildew and wheat leaf rust a defensin-like protein in tobacco styles. (PGR 99-071) Plant Physiol. 120, 663. pathogens. J. Basic Microbiol. 48, 516–520. McDonald, B. and Linde, C. (2002) The population genetics of plant pathogens and Broekaert, W., Cammue, B., De Bolle, M., Thevissen, K., De Samblanx, G. and breeding strategies for durable resistance. Euphytica, 124, 163–180. Osborne, R. (1997) Antimicrobial peptides from plants. Crit. Rev. Plant Sci. 16, Mygind, P., Fischer, R., Schnorr, K., Hansen, M., Sonksen, C., Ludvigsen, S., 297–323. Raventos, D., Buskov, S., Christensen, B., De Maria, L., Taboureau, O., Yaver, D., Bulet, P., Hetru, C., Dimarcq, J. and Hoffmann, D. (1999) Antimicrobial peptides in Elvig-Jorgensen, S., Sorensen, M., Christensen, B., Kjaerulff, S., insects; structure and function. Dev. Comp. Immunol. 23, 329–344. Frimodt-Moller, N., Lehrer, R., Zasloff, M. and Kristensen, H. (2005) Plectasin is Burdon, J.J. (1987) Disease and Plant Population Biology. London: Cambridge Univer- a peptide antibiotic with therapeutic potential from a saprophytic fungus. Nature, sity Press. 437, 975–980. Cawood, M., Pretorius, J., Westhuizen, A. and Pretorius, Z. (2010) Disease devel- Oldach, K., Becker, D. and Lörz, H. (2001) Heterologous expression of genes medi- opment and PR-protein activity in wheat (Triticum aestivum) seedlings treated with ating enhanced fungal resistance in transgenic wheat. Mol. Plant–Microbe Interact. plant extracts prior to leaf rust (Puccinia triticina) infection. Crop Prot. 29, 1311– 14, 832–838. 1319. Park, R., Fetch, T., Hodson, D., Jin, Y., Nazari, K., Prashar, M. and Pretorius, Z. Chang, T., Schroder, L.A., Thomson, J.M., Klocman, A.S., Tomasini, A.J., (2011) International surveillance of wheat rust pathogens—progress and challenges. Stromhaug, P.E. and Dunn, W.A., Jr (2005) PpATG9 encodes a novel membrane Euphytica, 1, 109–117. protein that traffics to vacuolar membranes, which sequester peroxisomes during Potter, L.R., Cagas, B., Paul, V.H. and Birckenstaedt, E. (1990) Pathogenicity of some pexophagy in Pichia pastoris. Mol. Biol. Cell 16, 4941–4953. European collections of crown rust (Puccinia coronata Corda) on cultivars of peren- Dangl, J. and Jones, J. (2001) Plant pathogens and integrated defense responses to nial ryegrass. J. Phytopathol. 130, 119–126. infection. Nature, 411, 826–833. Pretorius, Z., Singh, R., Wagoire, W. and Payne, T. (2000) Detection of virulence to De Coninck, B., Cammue, B. and Thevissen, K. (2013) Mode of antifungal action and wheat stem rust resistance gene Sr31 in Puccinia graminis f. sp. tritici in Uganda. in planta functions of plant defensins and defensin-like peptides. Fungal Biol. Rev. Plant Dis. 84, 203. 26, 109–120. Ramamoorthy, V., Zhao, X., Snyder, A., Xu, J.R. and Shah, D. (2007) Two mitogen- De Samblanx, G.W., Goderis, I.J., Thevissen, K., Raemaekers, R., Fant, F., activated protein kinase signalling cascades mediate basal resistance to antifungal Borremans, F., Acland, D.P., Osborn, R.W., Patel, S. and Broekaert, W.F. plant defensins in Fusarium graminearum. Cell. Microbiol. 9, 1491–1506. (1997) Mutational analysis of a plant defensin from radish (Raphanus sativus L.) Sagaram, U., Pandurangi, R., Kaur, J., Smith, T. and Shah, D. (2011) Structure– reveals two adjacent sites important for antifungal activity. J. Biol. Chem. 272, activity determinants in antifungal plant defensins MsDef1 and MtDef4 with differ- 1171–1179. ent modes of action against Fusarium graminearum. PLoS One 6, 1–13. Dracatos, P., Dumsday, J., Stewart, A., Dobrowolski, M., Cogan, N., Smith, K. and Schaaper, W.M., Posthuma, G.A., Plasman, H.H., Sijtsma, L., Fant, F., Borremans, Forster, J.W. (2008) Genetic diversity in Australasian populations of the crown rust F.A., Thevissen, K., Broekaert, W.F., Meloen, R.H. and van Amerongen, A. pathogen of ryegrasses (Puccinia coronata f. sp. lolii). In: Molecular Breeding of (2001) Synthetic peptides derived from the beta2-beta3 loop of Raphanus sativus Forage and Turf: The Proceedings of the 5th International Symposium on the Molecu- antifungal protein 2 that mimic the active site. J. Pept. Res. 57, 409–418. lar Breeding of Forage and Turf (Yamada, T. and Spangenberg, G., eds), pp. 275–285. Seale, S. and Vordtriede, P. (2010) Amino acid sequence variant alfalfa antifungal New York: Springer. protein and its use in plant disease control. Patent US 7728190 B1. Dracatos, P.M., Dobrowolski, M., Lamb, J., Olle, R., Gendall, A., Cogan, N., Smith, Silverstein, K., Graham, M., Paape, T. and VandenBosch, K. (2005) Genome organi- K. and Forster, J.W. (2009) Development of genetically homogenised populations of sation of more than 300 defensin-like genes in Arabidopsis. Plant Physiol. 138, the crown rust pathogen (Puccinia coronata f. sp. lolii) for disease trait dissection in 600–610. perennial ryegrass (Lolium perenne L.). Australas. Plant Pathol. 38, 55–62. Silverstein, K., Graham, M. and van den Bosch, K. (2006) Novel paralogous gene Dracatos, P.M., Keane, P.J. and Forster, J.W. (2010) Development of an efficient and families with potential function in legume nodules and seeds. Curr. Opin. Plant Biol. sensitive method for crown rust inoculation on ryegrasses in controlled environ- 9, 142–146. ments. Australas. Plant Pathol. 39, 562–570. Singh, R., Hodson, D., Jin, Y., Huerta-Espino, J., Kinyua, M., Wanyera, R., Njau, P. Gao, A., Hakimi, S., Mittanck, C., Wu, Y., Woerner, M., Stark, D., Shah, D., Liang, and Ward, R.W. (2006) Current status, likely migration and strategies to mitigate the J. and Rommens, C. (2000) Fungal pathogen protection in potato by expression of threat to wheat production from race Ug99 (TTKS) of stem rust pathogen. CAB Rev. a plant defensin peptide. Nat. Biotech. 18, 1307–1310. Perspect. Agric. Vet. Sci. Nutr. Nat. Resour. 1, 13. Ghag, S., Singh Shekhawat, U. and Ganapathi, T. (2012) Petunia floral defensins Singh, R., Hodson, D., Huerta-Espino, J., Jin, Y., Njau, P., Wanyera, R., with unique prodomains as novel candidates for development of fusarium wilt Herrera-Foessel, S. and Ward, R.W. (2008) Will stem rust destroy the worlds wheat resistance in transgenic banana plants. PLoS ONE 7, e39557. crop? Adv. Agron. 98, 71–80. Girgi, M., Breese, W., Lörz, H. and Oldach, K. (2006) Rust and downy mildew Spelbrink, R.G., Dilmac, N., Allen, A., Smith, T.J., Shah, D.M. and Hockerman, G.H. resistance in pearl millet (Pennisetum glaucum) mediated by heterologous expres- (2004) Differential antifungal and calcium channel-blocking activity among structur- sion of the afp gene from Aspergillus giganteus. Transgen. Res. 15, 313–324. ally related plant defensins. Plant Physiol. 135, 2055–2067.

Terras, F., Schoofs, H., De Bolle, M., van Leuven, F. and Rees, S. (1992) Analysis of Urdangarin, M., Norero, N., Broekaert, W. and de la Canal, L. (2000) A defensin two novel classes of plant antifungal proteins from radish (Raphanus sativus L.) gene expressed in sunflower inflorescence. Plant Phys. Biochem. 38, 253–258. seeds. J. Biol. Chem. 267, 15 301–15 309. van der Weerden, N. and Anderson, M. (2013) Plant defensins: common fold, Terras, F., Eggermont, K., Kovaleva, L., Raikhel, N., Osborn, R., Kester, A., Rees, S., multiple functions. Fungal Biol. Rev. 26, 121–131. Torrekens, S., Van Leuven, F., Vanderleyden, J., Cammue, B. and Broekaert, W. van der Weerden, N., Lay, F. and Anderson, M. (2008) The plant defensin NaD1 (1995) Small cysteine-rich antifungal proteins from radish: their role in host defense. enters the cytoplasm of Fusarium oxysporum hyphae. J. Biol. Chem. 283, 14 445– Plant Cell, 7, 573–588. 14 452. Thevissen, K., Cammue, B., Lemaire, K., Winderickx, J. and Dickson, R. (2000) A van der Weerden, N., Hancock, R. and Anderson, M. (2010) Permeabilisation of gene encoding a sphingolipid biosynthesis enzyme determines the sensitivity of fungal hyphae by the plant defensin NaD1 occurs through a cell wall-dependent Saccharomyces cerevisiae to an antifungal plant defensin from dahlia (Dahlia process. J. Biol. Chem. 285, 37 513–37 520. merckii). Proc. Natl. Acad. Sci. USA, 97, 9531–9536. Wiethölter, N., Horn, S., Reisige, K., Beike, U. and Moerschbacher, B. (2003) Thomma, B., Cammue, B. and Thevissen, K. (2002) Plant defensins. Planta, 216, In vitro differentiation of haustorial mother cells of the wheat stem rust fungus, 193–202. Puccinia graminis f. sp. tritici, triggered by the synergistic action of chemical and Tregear, J.W., Morcillo, F., Richaud, F., Berger, A., Singh, R., Cheah, S.C., physical signals. Fungal Genet. Biol. 38, 320–326. Hartmann, C., Rival, A. and Duval, Y. (2002) Characterization of a defensin gene Yount, N.Y. and Yeaman, M.R. (2004) Multidimensional signatures in antimicrobial expressed in oil palm inflorescences: induction during tissue culture and possible peptides. PNAS, 101, 7363–7368. association with epigenetic somaclonal variation events. J. Exp. Bot. 53, 1387– 1396.